implementation of drand48() as given in steve summit's book

[pereges]

I just tried to write a small program based on it :


#include <stdio.h>
#include <stdlib.h>

#define PRECISION 2.82e14

double drand48(void)
{
double x = 0;
double denom = RAND_MAX + 1;
double need;

for(need = PRECISION; need 1; need /= (RAND_MAX + 1.))
{
x += rand()/denom;
denom *= RAND_MAX + 1. ;
}

return x;

}


int main(void)
{
double x;
x = drand48();
printf("%f" ,x);
x = drand48();
printf("%f", x);

return 0;

}


But each time I get the same output 0. What could be wrong here ?
Does it mean that this will give teh same values over and over again
everytime its called ?

[jacob navia]

Re: implementation of drand48() as given in steve summit's book

pereges wrote:

I just tried to write a small program based on it :
>
>
#include <stdio.h>
#include <stdlib.h>
>
#define PRECISION 2.82e14
>
double drand48(void)
{
double x = 0;
double denom = RAND_MAX + 1;
double need;
>
for(need = PRECISION; need 1; need /= (RAND_MAX + 1.))
{
x += rand()/denom;
denom *= RAND_MAX + 1. ;
}
>
return x;
>
}
>
>
int main(void)
{
double x;
x = drand48();
printf("%f" ,x);
x = drand48();
printf("%f", x);
>
return 0;
>
}
>
>
But each time I get the same output 0. What could be wrong here ?
Does it mean that this will give teh same values over and over again
everytime its called ?

Using lcc-win I obtain

0.001268 0.585006

Of course I inserted a space after the first number.
It seems to be working

--
jacob navia
jacob at jacob point remcomp point fr
logiciels/informatique

http://www.cs.virginia.edu/~lcc-win32

[Eric Sosman]

Re: implementation of drand48() as given in steve summit's book

pereges wrote:

I just tried to write a small program based on it :
>
>
#include <stdio.h>
#include <stdlib.h>
>
#define PRECISION 2.82e14
>
double drand48(void)
{
double x = 0;
double denom = RAND_MAX + 1;

Are you sure this is copied correctly? As written here,
this is likely to misbehave if RAND_MAX == INT_MAX; to fix
that, change the final `1' to `1.0'.

--
Eric.Sosman@sun .com

[Barry Schwarz]

Re: implementation of drand48() as given in steve summit's book

On Mon, 21 Apr 2008 05:33:54 -0700 (PDT), pereges <Broli00@gmail. com>
wrote:

>I just tried to write a small program based on it :
>
>
>#include <stdio.h>
>#include <stdlib.h>
>
>#define PRECISION 2.82e14
>
>double drand48(void)
>{
> double x = 0;
> double denom = RAND_MAX + 1;
> double need;
>
> for(need = PRECISION; need 1; need /= (RAND_MAX + 1.))
{
> x += rand()/denom;
> denom *= RAND_MAX + 1. ;
}
>
> return x;
>
}
>
>
>int main(void)
>{
double x;
> x = drand48();
> printf("%f" ,x);
> x = drand48();
> printf("%f", x);
>
> return 0;
>
}
>
>
>But each time I get the same output 0. What could be wrong here ?

That's not the result I get. Did you cut and paste your code or
retype it?

>Does it mean that this will give teh same values over and over again
>everytime its called ?

Since your loop always executes the same number of times (four), the
return value is determined completely by the sequence of values
returned by rand. If you force rand to return the same sequence (such
as by calling srand with the same seed before each call to your
function), then you will get the same value each time. Otherwise not.

As it is now, since you don't call srand at all, your code will
produce the same sequence each time the program is run.


Remove del for email